The best quantitative measure of the integrity of a designed part, component, product or system

The probability that parts, components, products, or systems will perform their deigned-for functions without failure in specified environments for desired periods data given confidence level

Theoretical and practical tools

HALT testing is a process that is for discovery, specifically looking for failures

It is NOT a life test

The true value of a life parameter is never known, only it’s distribution about an expected value, so we cannot say when failure will occur. Sometimes we can say that the likelihood increases, but we can very rarely predict the time of failure.

Design Engineer

Product Engineer

Management

Customer or end user

It is important to remember that a HALT process should be continuously monitored, both by data acquisition equipment and by personnel

At least 3 samples as a minimum

Products likely to have be exposed to and possibly suffer from temperature, vibration, power cycling, or some combination

Your lab

Another test lab in your company

An outside test lab

Failure or the absence of failures is heavily dependent upon human actions and perceptions.

What may be a failure to one company may not be to another:

To get a better knowledge of your product

A single tool in the toolbox

To save ££££

The causes of failures that will occur in the future are often very uncertain

The probabilities of and durations to failures are also highly uncertain

Time spent on testing is expensive so the more quickly we can reduce the uncertainty gap the better

Finding causes of failure during development and preventing recurrence is far less expensive that finding new failure causes in use.

Should be started as soon as the hardware is available for test

For action to be taken to analyse and correct failure modes before production

As a way of tracking down causes of field failures

Actually A series of Tests

Temperature:

Operating Range

Shipping Range

Storage Range

Testing Range

Wrong question: At what stress do we decide that the level is so high that we can ignore the failures?

Right question: Could this failure occur in use?

Monitoring

Data Acquisition

Cabling Suitable for Test Environments

Talk with end users

Find out how they expect to use the vehicles

Get horror stories on the worst things that have ever happened

Find out the things that they have been most satisfied with

An engineer and an end user will look at the same product or system in very different ways.

Least destructive environment

Start at room ambient

LN2 will remove extra humidity

10 ° steps

Start with 10 degree steps.

Bring temperature up at the first sign of failure to see if it may be intermittent because of certain temperature.

Control the return to room temperature since this may cause it’s own failure.

Still Single Environment

5 ° C steps for assemblies

Know melting limits ahead of time

Start with 10 degree steps

Change to 5 degree steps as you get closer to suspected failure temperature (or melting point)

Bring temperature down at the first sign of a failure to see if it may be intermittent because of a certain temperature

Single Environment but multi-axis random

Start with 2 g steps

Keep temperature constant, preferably at laboratory ambient, so that vibration is your only stress

Bring g level down at the fist sign of a failure to see if it may be intermittent because of a certain acceleration

Hot and cold Test

5 or 10 °c steps depending on earlier results

Rapid ramps (thermal mismatch and conductivity effects)

Short dwells

Assuming a failure at 85 degrees we go 10 degrees lower

Give a 125 degree difference giving –50 as the cold range

According to Harry McLean in HALT HASS & HASA explained 3 to 5 cycles with 5 being preferred

According to Gregg K Hobbs in Accelerated Reliability Engineering he has not found any product to be rate sensitive so there is no need to limit the speed of the change rate

Heat

Vibration

Assuming a failure at 85 degrees we start at 65 ° giving a margin

Assuming a failure at 14 g’s we use 10 giving a margin

Vibration can be either pulsed or constant depending upon expected end usage

The same principal would work with a combination of cold and vibration

Thermal Swings:

Vibration

Use data based on baseline and earlier swing test

Assuming a failure at 50 g’s we divide that into fifths (because of five swings) and increase with each vibration burst

Humidity

Corrosion

Dust

Power Cycling

Remember that testing is a form of experimentation

Good guesses are valuable but need to be backed with real data

HALT is a way of getting a lot of data in a little time and for a comparatively small cost